Search

Because of spatial variability and complex compositions, the mechanical test results of natural soil–rock mixtures (SRMs) are often discrete and lack reproducibility, which has greatly restricted the practical application of the experimental findings. The objective of this study was to examine the general mechanical behavior of SRMs under the influences of some hidden factors (e.g., structural parameters, parent rock type and weathering degree). To that end, the abstraction idea was adopted to prepare purified SRM samples. Large-scale triaxial tests were performed on these purified materials. On this basis, the influences of three structural parameters on the mechanical behavior of SRMs were studied. Moreover, the relationship between the shear strength and parent rock type and that between the shear strength and the spatial distribution of rock blocks were quantified. Some additional intrinsic behavior was distinguished from individual experimental phenomena through the comparative analysis of the test data in this study and those reported in the literature. The results show that the hidden factors had significant influences on the mechanical behavior of SRMs. A greater saturated uniaxial compressive strength of rock blocks generally led to a larger shear strength of SRMs. According to the significance of their influences on the shear strength parameters of SRMs, the structural parameters are ordered as: the gradation of rock blocks, the initial dry density of sample and the spatial distribution of rock blocks. The deformation and failure feature of SRMs were considerably affected by the spatial distribution of rock blocks and shear rate. And the shear strength parameters of SRMs were mainly influenced by the content of grains between 40 and 60 mm. The findings of this study would provide useful guidance for engineering practice.

The durability of silty clay embankments is partially controlled by the moisture migration, which depends on soil hydraulic properties. This paper presents an experimental study of hydraulic properties of compacted binary silty clay. Specimens with different mixing ratios and dry densities were prepared. Scanning electron microscopy and mercury intrusion porosimetry were used to characterise the microstructure of silty clay. Thereafter, falling-head permeability tests and water retention tests were conducted to study the permeability and water retention property, respectively. The results demonstrate that clay particles are dispersed and show preferred arrangements after compaction when the clay content is 100%. As the clay content decreases, the arrangement of clay particles is gradually disturbed because of the existence of silt particles, causing the formation of large pores around silt particles. When the dry density increases, the pores around silt particles significantly decrease. Moreover, the permeability of silty clay decreases but the water retention capacity increases with increasing clay content and dry density. This is because the silty clay with larger clay content and dry density has fewer large pores, which greatly restrains the flow of water. Both the permeability and water retention property of silty clay can be predicted from pore size distribution parameters.

Ponded infiltration is very common in silty mudstone and has a great influence on the stability of related slopes, road cuttings, and tunnels. This paper aims to examine the infiltration performance of silty mudstone and predict the distribution of its water content under ponded conditions. By infiltration tests, the infiltration rate (i), cumulative infiltration (I), and their variations with the infiltration time (t) were obtained. Afterward, the variation of water content (w) with t and depth (s) was analyzed. The results show that the i value decreases with the increase in the degree of saturation, and the I value increases first significantly and then slightly during water infiltration. The entire w–t curve at any s is S-shaped, while the w–s curve at any t is full or half inverse-S-shaped. In addition, an equation was developed for the w–s prediction based on the simplified Gompertz curve model, and it was further extended to the spatial–temporal prediction model of water content. The evaluation results demonstrate that the spatial–temporal prediction model has high accuracy and reliability. The prediction model also indicates that the range of the infiltration-affected zone increases and the rate of increase slows down during water infiltration.

Based on a highway reconstruction project in southern China, this study aims to put forward a method to determine the proper treatment depth of the existing subgrade. First, some field tests including the Beckman beam deflection test and portable falling weight deflectometer (PFWD) test were carried out. The results showed that there was a good correlation between the Beckman beam deflection (L) and PFWD modulus (Ep). Subsequently, a subgrade section was excavated and backfilled with cement-stabilized soil in layers. Compaction test, dynamic cone penetrometer rate test, plate load test and Beckman beam deflection test were performed to evaluate the treatment effect. To make sure, the subgrade was treated deeply enough, the Beckman beam deflection (L) was used as the controlled indicator among all the measured indexes for it was the hardest metric to meet. According to the design deflection and decreasing law of the measured deflections with the different number of the stabilized layers, the treatment depth was finally determined. As the PFWD test is superior to the deflection test in the detection efficiency, and the deflection value can be calculated from PFWD modulus by correlation formulas, thus the latter index can be used as a more suitable parameter for estimating the treatment depth instead of the former. Consequently, based on the measured PFWD moduli of the existing subgrade, six treatment schemes considering different treatment depths were proposed. It was confirmed that the method developed from this study is feasible and worth being extensively applied.

Silty clay is a common embankment material whose moisture increases over time due to soil–atmosphere interactions and climate changes. The objective of this study is to investigate the uns...

Carbonaceous mudstone is a common embankment material with the characteristics of water softening and water disintegration, which often causes the instability of embankments. This paper pre...

The soil-water characteristic curve (SWCC) is often used to estimate unsaturated soil properties (e.g. strength, permeability, volume change, solute and thermal diffusivity). The SWCC of soil samples is significantly affected by cyclic wetting-drying. To examine how water content and cyclic wetting-drying affect the SWCC of disintegrated carbonaceous mudstone (DCM), SWCC tests were implemented using a pressure-plate apparatus. In addition, SWCC models for DCM considering the initial gravimetric water content and cyclic wetting-drying were developed. The test results showed that the volumetric water content (θ) of the DCM first decreased rapidly and then became stable as matric suction (s) increased. The initial water content affected the SWCC by altering the pore structure of the DCM. For a given number of wetting-drying cycles, the higher the initial water content, the higher the stabilized θ. At a given s value, θ decreased as the number of wetting-drying cycles increased, which suggests that cyclic wetting-drying reduces the water-holding capacity of DCM. The Gardner model for DCM was constructed considering initial water content and cyclic wetting-drying, and was effective at describing and predicting the SWCC model for DCM.

This work aimed to propose a semi-empirical model that predicts the permeability of saturated clay using the data of mercury intrusion porosimetry (MIP). First, the pore size distribution (PSD) curve obtained from an MIP test was regarded as a discrete probability function of pore diameters; thus, its shape could be characterized by probability parameters (e.g., the expected value and the standard deviation). Subsequently, these probability parameters, combined with the microporosity calculated from the volume of intruded mercury, were correlated with the permeability of clay based on Hagen-Poiseuille's equation. Next, the performance of the proposed permeability model was verified using data reported in the literature. Thereafter, the model was applied to estimate the permeability of normally consolidated (NC) and overconsolidated (OC) kaolin subjected to various triaxial loading. The results highlighted that the proposed model is capable of characterizing the sensitive variation of kaolin permeability under different overconsolidation ratios (OCRs), stress paths, and stress levels.

Wind is a dynamic variable, which produces rocking movements on wind turbines, and thus resulting in dynamic responses of foundations and the subsoil. This study attempts to examine the dynamic responses of the shallow foundation of an onshore wind turbine subjected to wind loads. To this end, a series of model tests considering different wind load conditions was carried out on a 1/10-scale physical model. The static responses, the dynamic responses and the dynamic amplification effects of the shallow foundation were subsequently measured and analysed. The results demonstrate that the dynamic responses of the wind turbine shallow foundation were visibly affected by both the mean wind load and the fluctuating wind load. At the rated wind speed, the maximum strain-based dynamic amplification factor (DAF) of the foundation baseplate reached as high as 1·65, the maximum deformation-based DAF of the foundation top was 1·14 and the maximum contact-pressure-based DAF on the downwind side was 1·14. In addition, the reliability of the physical model was discussed and the results were verified by comparison with the numerical data in the literature.

The presence of surface cracks has a great influence on the formation of transient saturated zones, which are one of the main factors affecting the stability of colluvial soil slopes during rainfall. This paper aims to examine how the distribution characteristics, such as depth (h), angle (θ), and location of surface cracks, influence the development of transient saturated zones. For this reason, a series of numerical simulations involving colluvial soil slopes is performed. The results show that the development of transient saturated zones in a slope with a crack and that in an intact slope are quite different. In a cracked slope, rainwater infiltrates into the slope along the crack and forms a transient saturated zone at the crack tip. It is found that the smaller the crack depth is, the easier it is for the transient saturated zone at the crack tip to connect with that in the shallow layer, leading to an increase in the area of the transient saturated zone in the shallow layer. Moreover, the larger the crack angle is, the more easily the transient saturated zone at the crack tip connects with both the transient saturated zone in the shallow layer and the groundwater, forming a transient saturated zone that penetrates the entire colluvium. In addition, the lower the location of the crack along the slope surface is, the shorter the time is for the formation of a transient saturated zone at the crack tip, and the larger the area of the transient saturated zone.

Slopes with carbonaceous mudstone (CM) are widely distributed in the southwest of China and have experienced numerous geological disasters in special climate, especially in rainfall conditions. Therefore, novel materials to stabilize CM slopes have attracted increasing interests. However, developing ultra-stable and cost-effective additives for CM slopes is still a great challenge. Herein, a hydrophobic polymeric material (polyvinylidene fluoride, PVDF) is investigated as an additive to enhance the mechanical strength and long-time stability of CM slopes. The PVDF is uniformly dispersed in CM matrix via interfacial interaction. The contact angle of the PVDF-modified carbonaceous mudstone (PVDF-MCM) can reach as high as 103.95°, indicating an excellent hydrophobicity. The unconfined compressive strength (UCS) and tensile strength (TS) of PVDF-MCM have been intensively enhanced to 4.07 MPa and 1.96 MPa, respectively, compared with ~0 MPa of pristine CM. Moreover, the UCS and TS of PVDF-MCM remain at 3.24 MPa and 1.03 MPa even after curing for 28 days in high humidity conditions. Our findings show that the PVDF can improve the hydrophobicity of CM significantly, which leads to super mechanical stability of PVDF-MCM. The excellent performance makes PVDF a promising additive for the development of ultra-stable, long-lifetime and cost-effective carbonaceous mudstone slopes.

International audience; This work aims to investigate the micropore properties of saturated kaolin subjected to triaxial loading. Consolidated drained triaxial compression tests were conducted on both normally consolidated and overconsolidated clay specimens. In the (p′, q) plane, a given stress level was reached by following two different stress paths (i.e., the purely deviatoric stress path and the conventional constant σ′3 stress path). This approach allows observing the influence of stress path on the evolution of pore structure. The experimental approach consisted in using scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP) to quantify the micropore structure of the specimens after being submitted to mechanical loading. The results clearly show that the micropore structure of clay was affected by the loading conditions. Consequently, the development of induced anisotropy appeared as mainly related to the variation of pore shape and pore orientation. Meanwhile, for overconsolidated specimens featuring dilative behavior, the pores exhibited a trend to be more open (i.e., rounder) and more randomly oriented compared with the normally consolidated specimens showing contractive behavior. Even if the micropore structures deduced from the SEM-based method, on the one hand, and from MIP tests, on the other hand, concern different scales of pores, the analysis shows that both methods of investigation were quite well comparable and revealed similar variation tendencies in void ratio and pore size distribution.

Gravelly soil is a typical heterogeneous porous medium with a multiscale structure and hydraulic conductivity that is challenging to quantify. The aim of this study is to examine the structure of an eluvial-colluvial gravelly soil at different scales and link the structural characteristics to the hydraulic conductivity. To this end, large gravelly soil samples and small fines-sand mixture samples were prepared and then characterized by X-ray computed tomography (CT) and optical microscopy, respectively. Through image analyses, the pore structural characteristics at the gravel scale (≥ 1.0 mm) and sand scale (0.01–1.0 mm) were identified. Constant-head tests were performed on the large gravelly soil samples to measure the saturated coefficients of permeability (k). The results show a relatively small gravel-scale porosity and a large sand-scale porosity in compacted gravelly soils. The dominant sizes of gravel-scale pores and sand-scale pores are several millimeters and approximately 0.06 mm, respectively. An evaluation of ten existing permeability equations indicates that most of the previous empirical equations proposed for sand and gravel are not applicable to well-graded gravelly soils. For this reason, the existing permeability equations were improved. In addition, novel empirical equations for estimating the k value of gravelly soils were proposed based on the structural parameters of pores, as well as the concepts of the effective porosity and the effective grain size.

Red clay, an easily cracking lateritic soil, is extensively distributed in southern China. Cracked red clay slopes often show shallow failures under rainfall, which is challenging to reproduce by finite element methods. This study aims to examine how desiccation cracks and vegetation affect the shallow stability of a red clay cut slope during rainfall. Numerical models of the slope are created using ABAQUS, and an effective approach is proposed to simulate the distribution of desiccation cracks. The reliability of the numerical models and approaches is validated against field surveys. Parametric analysis is then conducted to investigate the influences of rainfall parameters, crack parameters and vegetation parameters on the slope stability. The results demonstrate that the presence of desiccation cracks can cause shallow failure and greatly reduce the factor of safety of the slope. This indicates that control of crack development is important to maintain slope safety. Moreover, the stability of cracked slopes can be obviously improved by vegetation with roots longer than 0.5 m. A tap root system appears to perform better than a fibrous root system. However, shallow-rooted vegetation may also contribute to the shallow stability of the slope by restraining the formation of desiccation cracks.

Geotechnical problems in extremely complex environments have become increasingly prominent, which raises the need to explore more efficient solutions. Multidisciplinary approaches show a new perspective to deal with these problems. In this study, the electrochemical impedance spectroscopy (EIS) was introduced to illustrate the structural characteristics of soil-rock mixtures (SRM), aiming at deducing the relationship between physical property indices such as the mass ratio indices (e.g., water content, soil-rock ratio, and stone particle size) and volume ratio indices (e.g., porosity and saturation) and equivalent circuit at the macro and micro levels. A three-level equivalent circuit corresponding with the physical characteristic of SRM was established to describe the cementation state. It shows that the degree of cementation of SRM increases with increasing water content, rock content, and median size of block stone; meanwhile, the real impedance coefficient of the non-cemented part is lower than that of the cemented part. The corresponding relationship between the imaginary impedance and the specific gravity has segmental characteristics. The density first increases and then decreases with the increase of water content; the water content and saturation threshold are 16% and 80 ± 2%, respectively. Moreover, the saturation (Sr and n∙Sr) decreases with the increase in the real impedance coefficient. In addition, the characteristics of the propagation mode and interfacial contact characteristics of the charged material inside SRM were examined, and the macro and micro mapping of the cementation state of SRM and the basic mass transfer mode (i.e., migration, diffusion, and convection) were obtained. The mapping relationship proposed in this study reflects the micro-scale cementation characteristics and the macro-scale physical characteristics of the SRM system, which provides new insight into the multi-scale research of SRM.

Soil infiltration performances under rainfall affect shear strengths and play a crucial role in landslide problems. In this paper, the infiltration performances of gravelly soils under different rainfall conditions were investigated by 1D and 2D seepage tests. Computed tomography (CT) scans were adopted to identify the mesoscopic structural properties of gravelly soils. The significances of mesoscopic properties on the saturated coefficient of permeability (ks) were subsequently analyzed. The results showed a temporal and spatial variation of infiltration performances of a gravelly soil slope. The infiltration performances measured by 1D infiltration tests were different from those obtained from 2D tests. An exponential relationship between 1D and 2D infiltration rates was observed. The results further highlighted that void area ratio, uniformity coefficient, and gravel content were three most significant factors to ks. Three permeability formulas that respectively expressed by gravel content, initial dry density and two mesoscopic properties were proposed in this paper.

Wind loads are random variables, which induce significantly greater responses in structures than do static loads. We develop a finite-infinite element model of a 2 MW wind turbine using ABAQUS and then verify it with in situ data. The adopted dynamic constitutive model of the soil is based on the Davidenkov skeleton curve. The results demonstrate that the dynamic amplification factors (DAFs) strongly depend on wind speed and spatial position. Considerable values of the DAFs, ground acceleration, and ground velocity are observed, suggesting that the responses of the shallow foundation of a wind turbine are affected by dynamic wind loads.

This study aimed to investigate the influence of loosely bound water (LBW) on the compressibility of compacted fine-grained soils and accurately determine the soil’s compression index. Four fine-grained soils (i.e., heavy clay, heavy silt, lean clay, and lean silt) and a coarse-grained soil were examined. The volumetric flask method was used to measure the LBW contents of the five soils. X-ray diffraction (XRD) analysis was then performed to test the mineral compositions and help explain the reason why the LBW content varied between different soils. A concept of modified void ratio was proposed by assuming that LBW is a part of the solid phase in soil. Subsequently, consolidation tests and permeability tests were conducted on saturated compacted specimens. The results show that the compression indexes or permeability coefficients tend to be the same for the soils with identical initial modified void ratios. Consolidation tests were also carried out on the unsaturated compacted heavy silt of four different dry densities prepared at a water content higher than the optimum. They show that the compression of unsaturated soil occurs if pore air is discharged when the water content is less than the LBW content. This confirms the previous assumption that LBW can be regarded as a part of the soil solid phase. A modified compression index was deduced and implemented to predict the settlement of a road embankment. The result suggests that the modified compression index is capable of calculating the compression of fine-grained soils, whose water contents are higher than their LBW contents.

International audience; This experimental work aimed to identify local mechanisms in remoulded clays under different triaxial loading conditions. The behaviour at a given stress level reached by two different stress paths, i.e. the conventional constant σ′ 3 stress path and the purely deviatoric stress path, and the behaviour at two different stress levels along a given stress path were examined. The investigation concerned both normally consolidated and overconsolidated saturated specimens of remoulded kaolin clay. The microstructural states of the clay specimens induced by different triaxial loading paths were examined at two different scales using the scanning electron microscopy and the X-ray microtomography. The results highlighted five particle orientation modes that could be activated. These modes seem to be highly dependent on the stress level, the overconsolidation ratio and the stress path. Mesocracks were found to develop within strongly overconsolidated specimens accompanied by the occurrence of dilatancy. The proposed conceptual modes provide an interesting approach to understand the mechanical behaviour of remoulded clays, especially the macroscale dilatancy phenomenon, through the possible activation of different mechanisms at micro- and mesoscales.; Cette recherche expérimentale sur le comportement des argiles vise à tenter de comprendre, et d’identifier, les mécanismes qui s’activent au niveau local en lien avec les déformations volumiques sur chargements triaxiaux. La démarche expérimentale consiste à mettre en évidence l’influence du chemin de chargement sur le comportement micro-macro. Deux différents chemins triaxiaux, l’un conventionnel à σ’ 3 constante et l’autre purement déviatoire à p’ constant, ont été menés pour atteindre le même niveau de contrainte fixé sur le plan des contraintes. Les essais ont été réalisés sur des échantillons d’argile saturée remaniée reconstituée au laboratoire, en condition normalement consolidée et surconsolidée. L’investigation microstructurale a été menée à deux échelles différentes : au moyen de la Microscopie Electronique à Balayage (MEB), et de la Microtomographie. Les résultats ont mis en évidence l’activation possible de cinq modes d’orientation privilégiée des particules argileuses. Ces modes d’orientation sont fortement dépendants du niveau de contrainte et de l’histoire du chargement. Un développement d’une microfissuration a été observé à l’échelle microscopique et mésoscopique dans les échantillons fortement surconsolidés. L’évolution de cette micro (mais aussi et méso) fissuration semble accompagner clairement le phénomène de la dilatance. Les modes conceptuels d’orientation privilégiée avancés dans ce travail, permettent de mieux comprendre le phénomène de dilatance et de mettre en lumière les mécanismes qui le gouvernent.

International audience; This experimental work aimed to identify local mechanisms in remoulded clays under different triaxial loading conditions. The behaviour at a given stress level reached by two different stress paths, i.e. the conventional constant σ′3 stress path and the purely deviatoric stress path, and the behaviour at two different stress levels along a given stress path were examined. The investigation concerned both normally consolidated and overconsolidated saturated specimens of remoulded kaolin clay. The microstructural states of the clay specimens induced by different triaxial loading paths were examined at two different scales using the scanning electron microscopy and the X-ray microtomography. The results highlighted five particle orientation modes that could be activated. These modes seem to be highly dependent on the stress level, the overconsolidation ratio and the stress path. Mesocracks were found to develop within strongly overconsolidated specimens accompanied by the occurrence of dilatancy. The proposed conceptual modes provide an interesting approach to understand the mechanical behaviour of remoulded clays, especially the macroscale dilatancy phenomenon, through the possible activation of different mechanisms at micro- and mesoscales.

This study aims to examine the mechanical behavior of disintegrated carbonaceous mudstone modified with nanosilica and cement (DCMNC). Many DCMNC specimens with various nanosilica contents were prepared. The X-ray diffraction (XRD) analyses and scanning electron microscopy (SEM) observations were performed on some of the specimens. Afterwards, triaxial tests were carried out on the remaining specimens to determine the mechanical behavior of DCMNC. The results showed that the cohesion exhibited a positive correlation with nanosilica content while the angle of internal friction presented a negative correlation with nanosilica content. The peak deviatoric stress, residual deviatoric stress and brittle modulus of DCMNC showed an increase followed by a decrease as nanosilica content varied from 0 to 8%, and all of them reached corresponding maximums at a nanosilica content of 2%. Thus, 2% was considered to be the optimum nanosilica content. The modification mechanism of DCMNC could be explained by the pozzolanic reaction related to nanosilica and the filling effect of nanosilica.

This study aims to examine the deformation behavior and internal mechanism of coarse-grained soil as an embankment filler under cyclic loading. Numerical dynamic triaxial tests were performed on coarse-grained soil using the discrete element software PFC3D. The numerical model was verified by comparing the numerical results with the experimental data. Afterward, the changes in the porosity, force chain, and particle movement of coarse-grained soil samples were analyzed, and the mesoscopic deformation behavior of coarse-grained soil under cyclic loading was investigated. The research results show that with the increase of the deviatoric stress amplitude, moisture content, and loading frequency, the deformation of the soil increases and the ability to resist deformation decreases at the same loading cycles. Due to the inhomogeneous distribution of particles with different sizes, the velocity and displacement of the sample vary in different directions, exhibiting mesoscopic anisotropy. The contact force is relatively even in the downward direction while dispersed near the edge of the sample. This means that the particles at the bottom are less affected by loads and the internal evolution of soil samples conforms to its macroscopic deformation behavior during cyclic loading.

Seasonal rainfall causes a carbonaceous mudstone embankment to undergo cyclic drying/wetting. The soil mass inside the embankment is within a mechanically loaded and soaking environment, which results in a significant degradation of the shear strength of the soil and excessive deformation. Herein, to study the deformation and shear strength characteristics of disintegrated carbonaceous mudstone (DCM), a new type of wetting deformation test device was developed. Wetting deformation tests were carried out on the basis of an orthogonal experimental design, and direct shear tests and computed tomography (CT) scans were then conducted on the wetting deformed specimens. The test results show that the DCM exhibits a large vertical deformation at the time of the initial loading and water infiltration, termed compressive deformation and wetting deformation, respectively. The change in the shear strength during the wetting deformation is mainly due to the change in cohesion. The shear strength decreases sharply and then tends to stabilize with the increasing vertical deformation. The results of CT scans reveal that the changes in the shear strength and deformation of the DCM are associated with changes in the pore structure inside the specimen.

The objective of this study was to examine the fracture evolution and related mechanism of silty mudstone in cyclic hydrothermal environments. Fracture evolution tests and computed tomography (CT) scans were carried out on silty mudstone specimens subjected to cyclic wetting-heating. The spatial–temporal variations of fracture parameters including the total length ratio, fracture rate, mean fracture width, fracture direction and fractal dimension were investigated. The mechanism underlying the fracture evolution in silty mudstone was also analyzed. The results show that fractures initiate and propagate from the wetting-heating surface down to the deep layer with increasing cycle number. Most of the fractures are parallel or perpendicular to each other and form a reticulate fracture network on every horizontal cross-section. The total length ratio, fracture rate, mean fracture width and fractal dimension considerably decrease as the depth increases, and the total length ratio and fracture rate significantly increase with increasing cycle number. After a wetting-heating cycle, only the zone within a depth of 14 mm shows slight fracturing; however, after 17 wetting-heating cycles, slight fracturing occurs within a depth of 104 mm. The fracture mechanism mainly involves the hydrothermal stress effect, water-softening effect, and mineral hydrolysis.

International audience; The aim of this study is to understand the local mechanisms related to creep behavior of a typical clay under triaxial loading. The investigation concerns both normally consolidated and overconsolidated specimens of remolded kaolin clay. The macroscopic results showed that both dilatancy and contractancy could occur during creep depending mainly on the clay behavior prior to the creep test. The magnitude of the dilatancy/contractancy was controlled by the stress level, on one hand, and by the overconsolidation ratio which governed the sign of the volumetric strain variation during triaxial loading, on the other hand. At the microscopic scale, the dilatancy/contractancy phenomena were analyzed using the scanning electron microscopy (SEM). The results indicated that the microstructural evolution of the clay along mechanical loading depended on the stress history. The structural evolution during the creep phases followed the structural pattern developed under monotonic loading. The creep dilatancy phenomenon appeared strongly related to the expansion of micro pores and micro cracks within the overconsolidated clay specimens.

Completely disintegrated carbonaceous mudstone, a common embankment material, has significant swelling-shrinkage behavior under cyclic wetting and drying, which often causes the instability of embankments. In this paper, the evolution of the cracks in the completely disintegrated carbonaceous mudstone subjected to cyclic wetting and drying was studied by laboratory tests. The vacuum saturation method and the drying method based on heat lamps and a fan were employed to simulate the wetting and drying processes, respectively. The image processing technique was used to treat the images of the sample surface. Afterward, various geometric parameters of the cracks appeared on the sample surface were measured, and the evolution characteristics of the cracks were analyzed. The results show that with the increase in the number of wetting and drying cycles, the number of cracks on the sample surface gradually increases. After four wetting and drying cycles, the existing cracks produce a large number of small branches, which are connected to form irregular polygonal grids. The development of the cracks in the completely disintegrated carbonaceous mudstone sample can be divided into three stages, i.e., slow development, rapid development, and stable development. Both the surface density of crack and the crack rate first increase and then tend to be stable with the increasing number of wetting and drying cycles. New cracks are mainly generated on the basis of the existing cracks, and some cracks appear to heal under cyclic wetting and drying. The early cracks are mainly developed in the directions of 0°–90° and 300°–360°, and the development rate of the cracks in the directions of 30°, 90°, 140°–150°, and 180°–270° is significantly higher than that of the cracks in other directions.

International audience; The aim of this study is to analyze the influence of stress path on the dilatancy behavior of clays. Triaxial tests were conducted on saturated, remolded clay specimens. During triaxial loading, two different stress paths were considered to bring the stress level to a given point in the (p' - q) plane. After triaxial testing, scanning electron microscopy (SEM) observations and X-ray microtomography (XR-μCT) scans were performed on subsamples. The obtained images were processed using different software and thus the geometric characteristics of pores and cracks were identified. The microstructural characteristics were linked to the dilatancy phenomenon of specimens. The results indicate that dilatancy is influenced by stress path and may be attributed to the evolution of pore geometry and the presence of local open cracks in highly over-consolidated (OC) clays..

The stability of a residual gravelly soil slope during heavy rainfall is closely related to the seepage field, which is directly affected by the seepage boundaries of the slope. To examine the boundary effect, a novel model flume that can implement permeable and impermeable boundary conditions was designed. With this flume, hydraulic model tests of a gravelly soil slope considering those two seepage boundaries were carried out. The results show that distinct spatial morphology of seepage fields is reflected by the hydraulic characteristics of the slope models with different seepage boundaries. The slope with permeable boundaries generally exhibits a larger average infiltration rate as well as a relatively smaller moving speed of wetting front in the lower part compared to the slope with impermeable boundaries. The residual moisture content in the deep-lower part of the slope with permeable boundaries is 6.6% cm3/cm3 smaller than that of the slope with impermeable boundaries. Nevertheless, the maximum pore water pressure in the middle part of the slope with permeable boundaries is 1.2 kPa higher than that in the corresponding part of the slope with impermeable boundaries. It is noted that regional pore water pressures in the shallow-lower and shallow-middle parts of the slope with impermeable boundaries exhibit large fluctuations with an amplitude over 0.5 kPa. The seepage field in the slope with permeable boundaries presents a multi-dimensional development, which may lead to a local failure of the slope. The results are of great significance in studying the similarity of seepage boundary condition in the field of similarity theory on landslide model tests and provide an experimental basis for establishing rainfall-induced landslide theory.